## Load libraries
library(covid19)
library(ggplot2)
library(lubridate)
library(dplyr)
library(ggplot2)
library(sp)
library(raster)
library(viridis)
library(ggthemes)
library(sf)
library(rnaturalearth)
library(rnaturalearthdata)
library(RColorBrewer)
pd <- df_country
pd$value <- pd$deaths_non_cum
pd <- pd %>%
filter(date == max(date)) %>%
dplyr::select(country, iso, cases, cases_non_cum,
deaths, value) %>%
dplyr::arrange(desc(value)) %>%
left_join(world_pop %>% dplyr::select(-country)) %>%
mutate(value_per_million = value / pop * 1000000) #%>%
# arrange(desc(value_per_million))
pd <- pd[1:10,]
pd$country <- factor(pd$country, levels = pd$country)
ggplot(data = pd,
aes(x = country,
y = value)) +
geom_bar(stat = 'identity',
fill = 'black') +
theme_simple() +
geom_text(aes(label = value),
nudge_y = -20,
size = 4,
color = 'white')
pd
# A tibble: 10 x 10
# Groups: country [10]
country iso cases cases_non_cum deaths value pop region sub_region
<fct> <chr> <dbl> <dbl> <dbl> <dbl> <dbl> <chr> <chr>
1 Spain ESP 64059 7871 4858 769 4.67e7 Europe Southern …
2 Italy ITA 80539 6153 8165 662 6.04e7 Europe Southern …
3 France FRA 29551 3951 1698 365 6.70e7 Europe Western E…
4 US USA 83836 18058 1209 267 3.27e8 Ameri… Northern …
5 Iran IRN 29406 2389 2234 157 8.18e7 Asia Southern …
6 United… GBR 11812 2172 580 114 6.65e7 Europe Northern …
7 Nether… NLD 7468 1030 435 78 1.72e7 Europe Western E…
8 Germany DEU 43938 6615 267 61 8.29e7 Europe Western E…
9 North … PRK 7513 7513 54 54 2.55e7 Asia Eastern A…
10 Belgium BEL 6235 1298 220 42 1.14e7 Europe Western E…
# … with 1 more variable: value_per_million <dbl>
Deaths per million yesterday per CCAA
pd <- esp_df
pd$value <- pd$deaths_non_cum
pd <- pd %>%
filter(date == max(date)) %>%
dplyr::select(ccaa, cases, cases_non_cum,
deaths, value) %>%
dplyr::arrange(desc(value)) %>%
left_join(esp_pop)%>%
mutate(value_per_million = value / pop * 1000000) #%>%
# arrange(desc(value_per_million))
pd <- pd[1:10,]
pd$country <- factor(pd$country, levels = pd$country)
Error in `$<-.data.frame`(`*tmp*`, country, value = structure(integer(0), .Label = character(0), class = "factor")): replacement has 0 rows, data has 10
ggplot(data = pd,
aes(x = country,
y = value)) +
geom_bar(stat = 'identity',
fill = 'black') +
theme_simple() +
geom_text(aes(label = value),
nudge_y = -20,
size = 4,
color = 'white')
Error in FUN(X[[i]], ...): object 'country' not found
pd
# A tibble: 10 x 7
ccaa cases cases_non_cum deaths value pop value_per_million
<chr> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
1 Madrid 19243 2077 2412 322 6663394 48.3
2 Cataluña 12940 1348 880 208 7675217 27.1
3 CLM 3934 551 367 51 2032863 25.1
4 CyL 4132 644 252 46 2399548 19.2
5 C. Valenciana 3532 332 198 31 5003769 6.20
6 País Vasco 4601 655 207 27 2207776 12.2
7 La Rioja 1236 241 55 12 316798 37.9
8 Extremadura 1231 262 69 11 1067710 10.3
9 Galicia 2322 407 43 11 2699499 4.07
10 Andalucía 3793 387 144 10 8414240 1.19
dir.create('~/Desktop/animation_deaths')
dates <- seq(as.Date('2020-03-17'), (Sys.Date()-1), by = 1)
for(i in 1:length(dates)){
this_date <- dates[i]
pd <- df_country
pd$value <- pd$deaths_non_cum
pd <- pd %>%
filter(date == max(this_date)) %>%
dplyr::select(country, cases, cases_non_cum,
deaths, value) %>%
dplyr::arrange(desc(value))
pd <- pd[1:10,]
pd <- pd %>% filter(value > 0)
pd$country <- gsub(' ', '\n', pd$country)
pd$country <- factor(pd$country, levels = pd$country)
pd$color_var <- pd$country == 'Spain'
if('Spain' %in% pd$country){
cols <- rev(c('darkred', 'black'))
} else {
cols <- 'black'
}
g <- ggplot(data = pd,
aes(x = country,
y = value)) +
geom_bar(stat = 'identity',
aes(fill = color_var),
alpha = 0.8,
show.legend = FALSE) +
theme_simple() +
geom_text(aes(label = value),
nudge_y = max(pd$value) * .05,
size = 5,
color = 'black') +
labs(title = 'Daily (non-cumulative) COVID-19 deaths',
subtitle = format(this_date, '%B %d')) +
labs(x = 'Country',
y = 'Deaths') +
theme(axis.text = element_text(size = 15),
plot.subtitle = element_text(size = 20)) +
scale_fill_manual(name ='',
values = cols) +
ylim(0, 900)
ggsave(filename = paste0('~/Desktop/animation_deaths/', this_date, '.png'),
g)
}
# Command line
cd ~/Desktop/animation_deaths
mogrify -resize 50% *.png
convert -delay 50 -loop 0 *.png result.gif
pd <- by_country <- esp_df %>% mutate(country = 'Spain') %>% bind_rows(ita %>% mutate(country = 'Italy'))
pd$value <- pd$deaths_non_cum
max_date <- pd %>% group_by(country) %>% summarise(d = max(date)) %>% ungroup %>% summarise(d = min(d)) %>% .$d
# pd$value <- ifelse(is.na(pd$value), 0, pd$value)
left <- expand.grid(date = seq(min(pd$date),
max(pd$date),
by = 1),
ccaa = sort(unique(pd$ccaa)))
right <- pd %>% dplyr::select(date, ccaa, value)
pd <- left_join(left, right) %>% mutate(value = ifelse(is.na(value), NA, value))
pd <- left_join(pd, by_country %>% dplyr::distinct(country, ccaa)) %>%
filter(date <= max_date) %>%
filter(value > 0)
the_limits <- c(1, 600)
the_breaks <- c(1, seq(100, 600, length = 6)) #seq(0, 600, length = 7)
pd$ccaa <- factor(pd$ccaa, levels = rev(unique(sort(pd$ccaa))))
ggplot(data = pd,
aes(x = date,
y = ccaa,
color = value,
size = value)) +
# geom_tile(color = 'white') +
geom_point(alpha = 0.8) +
scale_color_gradientn(colors = rev(colorRampPalette(brewer.pal(n = 8, 'Spectral'))(5)),
name = '',
limits = the_limits,
breaks = the_breaks) +
scale_size_area(name = '', limits = the_limits, breaks = the_breaks, max_size = 10) +
theme_simple() +
facet_wrap(~country, scales = 'free') +
theme(strip.text = element_text(size = 26),
axis.title = element_blank(),
axis.text = element_text(size = 16)) +
guides(color = guide_legend(), size = guide_legend()) +
labs(title = 'Daily (non-cumulative) COVID-19 deaths by sub-state regions',
caption = paste0('Data as of ', max_date))
ggsave('/tmp/1.png')
pd <- by_country <- esp_df %>% mutate(country = 'Spain') %>% bind_rows(ita %>% mutate(country = 'Italy'))
poppy <- bind_rows(ita_pop, esp_pop)
pd <- pd %>% left_join(poppy)
pd$value <- pd$deaths_non_cum / pd$pop * 1000000
max_date <- pd %>% group_by(country) %>% summarise(d = max(date)) %>% ungroup %>% summarise(d = min(d)) %>% .$d
# pd$value <- ifelse(is.na(pd$value), 0, pd$value)
left <- expand.grid(date = seq(min(pd$date),
max(pd$date),
by = 1),
ccaa = sort(unique(pd$ccaa)))
right <- pd %>% dplyr::select(date, ccaa, value)
pd <- left_join(left, right) %>% mutate(value = ifelse(is.na(value), NA, value))
pd <- left_join(pd, by_country %>% dplyr::distinct(country, ccaa)) %>%
filter(date <= max_date) %>%
filter(value > 0)
the_limits <- c(1, 60)
the_breaks <- c(1, seq(10, 60, length = 6)) #seq(0, 600, length = 7)
pd$ccaa <- factor(pd$ccaa, levels = rev(unique(sort(pd$ccaa))))
ggplot(data = pd,
aes(x = date,
y = ccaa,
color = value,
size = value)) +
# geom_tile(color = 'white') +
geom_point(alpha = 0.8) +
scale_color_gradientn(colors = rev(colorRampPalette(brewer.pal(n = 8, 'Spectral'))(5)),
name = '',
limits = the_limits,
breaks = the_breaks) +
scale_size_area(name = '', limits = the_limits, breaks = the_breaks, max_size = 10) +
theme_simple() +
facet_wrap(~country, scales = 'free') +
theme(strip.text = element_text(size = 26),
axis.title = element_blank(),
axis.text = element_text(size = 16)) +
guides(color = guide_legend(), size = guide_legend()) +
labs(title = 'Daily COVID-19 deaths per 1,000,000 population by sub-state regions',
caption = paste0('Data as of ', max_date))
ggsave('/tmp/2.png')
place_transform <- function(x){ifelse(x == 'Madrid', 'Madrid',
# ifelse(x == 'Cataluña', 'Cataluña',
'Otras CCAA')
# )
}
place_transform_ita <- function(x){
ifelse(x == 'Lombardia', 'Lombardia',
# ifelse(x == 'Emilia Romagna', 'Emilia Romagna',
'Otras regiones italianas')
# )
}
pd <- esp_df %>% mutate(country = 'España') %>%
mutate(ccaa = place_transform(ccaa)) %>%
bind_rows(ita %>% mutate(ccaa = place_transform_ita(ccaa),
country = 'Italia')) %>%
group_by(country, ccaa, date) %>%
summarise(cases = sum(cases),
uci = sum(uci),
deaths = sum(deaths),
cases_non_cum = sum(cases_non_cum),
deaths_non_cum = sum(deaths_non_cum),
uci_non_cum = sum(uci_non_cum)) %>%
left_join(esp_pop %>%
mutate(ccaa = place_transform(ccaa)) %>%
bind_rows(ita_pop %>% mutate(ccaa = place_transform_ita(ccaa))) %>%
group_by(ccaa) %>%
summarise(pop = sum(pop))) %>%
mutate(deaths_non_cum_p = deaths_non_cum / pop * 1000000) %>%
group_by(country, date) %>%
mutate(p_deaths_non_cum_country = deaths_non_cum / sum(deaths_non_cum) * 100,
p_deaths_country = deaths / sum(deaths) * 100)
pd$ccaa <- factor(pd$ccaa,
levels = c('Madrid',
# 'Cataluña',
'Otras CCAA',
'Lombardia',
# 'Emilia Romagna',
'Otras regiones italianas'))
cols <- c(
rev(brewer.pal(n = 3, 'Reds'))[1:2],
rev(brewer.pal(n = 3, 'Blues'))[1:2]
)
ggplot(data = pd,
aes(x = date,
y = deaths_non_cum_p,
fill = ccaa,
group = ccaa)) +
geom_bar(stat = 'identity',
position = position_dodge(width = 0.99)) +
scale_fill_manual(name = '', values = cols) +
scale_color_manual(name = '', values = cols) +
# geom_line(size = 0.2,
# aes(color = ccaa)) +
xlim(as.Date('2020-03-09'),
Sys.Date()-1) +
theme_simple() +
labs(x = 'Fecha',
y = 'Muertes diarias por 1.000.000',
title = 'Muertes por 1.000.000 habitantes') +
theme(legend.position = 'top') +
geom_text(aes(label = round(deaths_non_cum_p, digits = 1),
color = ccaa,
y = deaths_non_cum_p + 2,
group = ccaa),
size = 2.5,
position = position_dodge(width = 0.99))
label_data <- pd %>%
filter(country %in% c('Italia', 'España')) %>%
group_by(country) %>%
filter(date == max(date)) %>%
mutate(label = gsub('\nitalianas', '', gsub(' ', '\n', ccaa))) %>%
mutate(x = date - 2,
y = p_deaths_country +
ifelse(p_deaths_country > 50, 10, -12))
ggplot(data = pd %>% group_by(country) %>% mutate(start_day = dplyr::first(date[deaths >=50])) %>% filter(date >= start_day),
aes(x = date,
y = p_deaths_country,
color = ccaa,
group = ccaa)) +
# geom_bar(stat = 'identity',
# position = position_dodge(width = 0.99)) +
scale_fill_manual(name = '', values = cols) +
scale_color_manual(name = '', values = cols) +
geom_line(size = 2,
aes(color = ccaa)) +
geom_point(size = 3,
aes(color = ccaa)) +
facet_wrap(~country, scales = 'free_x') +
# xlim(as.Date('2020-03-09'),
# Sys.Date()-1) +
theme_simple() +
geom_hline(yintercept = 50, lty = 2, alpha = 0.6) +
# geom_line(lty = 2, alpha = 0.6) +
labs(x = 'Fecha',
y = 'Porcentaje de muertes',
title = 'Porcentaje de muertes diarias atribuible a la región más afectada vs. resto del país',
subtitle = 'A partir del primer día en cada país con 50 o más muertes') +
theme(legend.position = 'top',
strip.text = element_text(size = 30),
legend.text = element_text(size = 10)) +
guides(color = guide_legend(nrow = 2,
keywidth = 5)) +
geom_text(data = label_data,
aes(x = x,
y = y,
label = label,
color = ccaa),
size = 7,
show.legend = FALSE)
ggsave('~/Desktop/porcentaje.png')
# Spanish data
a <- esp_df %>%
left_join(esp_pop) %>%
mutate(country = 'Spain')
# Italian data
b <- ita %>%
left_join(ita_pop) %>%
mutate(country = 'Italy')
# Chinese data
d <- df %>% filter(country == 'China') %>%
mutate(cases = cases) %>%
mutate(ccaa = district) %>%
mutate(country = 'China') %>%
left_join(chi_pop)
# join
joined <- bind_rows(a, b, d)
# Get deaths per milllion
joined$deaths_pm <- joined$deaths / joined$pop * 1000000
joined$cases_pm <- joined$cases / joined$pop * 1000000
# Get the days since paradigm
x_deaths <- 5
x_deaths_pm <- 5
x_cases <- 50
x_cases_pm <- 50
joined <- joined %>%
arrange(date) %>%
group_by(ccaa) %>%
mutate(start_deaths = min(date[deaths >= x_deaths]),
start_cases = min(date[cases >= x_cases]),
start_deaths_pm = min(date[deaths_pm >= x_deaths_pm]),
start_cases_pm = min(date[cases_pm >= x_cases_pm])) %>%
ungroup %>%
mutate(days_since_start_deaths = date - start_deaths,
days_since_start_cases = date - start_cases,
days_since_start_deaths_pm = date - start_deaths_pm,
days_since_start_cases_pm = date - start_cases_pm)
# Define plot data
pd <- joined %>% filter(days_since_start_deaths_pm >= 0) %>%
mutate(country = ifelse(country == 'China',
'Hubei (China)',
ifelse(country == 'Italy', 'Italia', 'España')))
# Define label data
label_data <- pd %>% group_by(ccaa) %>% filter(
(
(country == 'Hubei (China)' & days_since_start_deaths_pm == 22) |
(date == max(date) & country == 'España' & deaths_pm > 40 & days_since_start_deaths_pm >= 7) & ccaa != 'CyL' |
(date == max(date) & country == 'Italia' &
ccaa != 'Liguria' & days_since_start_deaths_pm > 15)
))
# Get differential label data based on what to be emphasized
bigs <- c('Madrid', 'Lombardia', 'Hubei')
label_data_big <- label_data %>%
filter(ccaa %in% bigs)
label_data_small <- label_data %>%
filter(!ccaa %in% bigs)
pd_big <- pd %>%
filter(ccaa %in% bigs)
pd_small <- pd %>%
filter(!ccaa %in% bigs)
# cols <- colorRampPalette(RColorBrewer::brewer.pal(n = 8, name = 'Set2'))(length(unique(pd$country)))
# cols <- rainbow(3)
cols <- c('darkred', '#FF6633', '#006666')
ggplot(data = pd_big,
aes(x = days_since_start_deaths_pm,
y = deaths_pm,
group = ccaa)) +
geom_line(aes(color = country),
alpha = 0.9,
size = 2) +
geom_line(data = pd_small,
aes(x = days_since_start_deaths_pm,
y = deaths_pm,
color = country),
alpha = 0.7,
size = 1) +
scale_y_log10() +
scale_color_manual(name = '',
values = c(cols)) +
theme_simple() +
theme(legend.position = 'top') +
labs(x = 'Dias desde "el comienzo del brote"',
y = 'Muertes por millón de habitantes',
title = 'Muertes por 1.000.000 habitantes',
subtitle = paste0('Dia 0 ("comienzo del brote") = primer día a ', x_deaths_pm, ' o más muertes acumuladas por milión de población\nLíneas rojas: CCAA; líneas verde-azules: regiones italianas; línea naranja: Hubei, China'),
caption = '@joethebrew | www.databrew.cc') +
geom_text(data = label_data_big,
aes(x = days_since_start_deaths_pm + 0.6,
y = (deaths_pm + 50),
label = gsub(' ', '\n', ccaa),
color = country),
size = 8,
alpha = 0.9,
show.legend = FALSE) +
geom_text(data = label_data_small,
aes(x = days_since_start_deaths_pm + 0.6,
y = deaths_pm + (log(deaths_pm)/2),
label = gsub(' ', '\n', ccaa),
color = country),
size = 5,
alpha = 0.6,
show.legend = FALSE) +
theme(axis.text = element_text(size = 14),
axis.title = element_text(size = 20),
legend.text = element_text(size = 16),
plot.title = element_text(size = 30)) +
xlim(0, 26)
ggsave('~/Desktop/china_spain_italy_comparison.png',
height = 7,
width = 10)
# Spanish data
a <- esp_df %>%
left_join(esp_pop) %>%
mutate(country = 'Spain')
# Italian data
b <- ita %>%
left_join(ita_pop) %>%
mutate(country = 'Italy')
# Chinese data
d <- df %>% filter(country == 'China') %>%
mutate(cases = cases) %>%
mutate(ccaa = district) %>%
mutate(country = 'China') %>%
left_join(chi_pop)
# join
joined <- bind_rows(a, b, d)
# Get deaths per milllion
joined$deaths_pm <- joined$deaths / joined$pop * 1000000
joined$cases_pm <- joined$cases / joined$pop * 1000000
# Get the days since paradigm
x_deaths <- 5
x_deaths_pm <- 5
x_cases <- 50
x_cases_pm <- 50
joined <- joined %>%
arrange(date) %>%
group_by(ccaa) %>%
mutate(start_deaths = min(date[deaths >= x_deaths]),
start_cases = min(date[cases >= x_cases]),
start_deaths_pm = min(date[deaths_pm >= x_deaths_pm]),
start_cases_pm = min(date[cases_pm >= x_cases_pm])) %>%
ungroup %>%
mutate(days_since_start_deaths = date - start_deaths,
days_since_start_cases = date - start_cases,
days_since_start_deaths_pm = date - start_deaths_pm,
days_since_start_cases_pm = date - start_cases_pm)
# Define plot data
pd <- joined %>% filter(days_since_start_deaths >= 0) %>%
mutate(country = ifelse(country == 'China',
'China',
ifelse(country == 'Italy', 'Italia', 'España')))
# Define label data
label_data <- pd %>% group_by(ccaa) %>% filter(
(
(country == 'China' & deaths >10 & days_since_start_deaths == 29) |
(date == max(date) & country == 'España' & deaths > 90) |
(date == max(date) & country == 'Italia' &
ccaa != 'Liguria' & days_since_start_deaths > 10)
))
# Get differential label data based on what to be emphasized
label_data_big <- label_data %>%
filter(ccaa %in% c('Madrid', 'Lombardia', 'Hubei'))
label_data_small <- label_data %>%
filter(!ccaa %in% c('Madrid', 'Lombardia', 'Hubei'))
pd_big <- pd %>%
filter(ccaa %in% c('Madrid', 'Lombardia', 'Hubei'))
pd_small <- pd %>%
filter(!ccaa %in% c('Madrid', 'Lombardia', 'Hubei'))
# cols <- colorRampPalette(RColorBrewer::brewer.pal(n = 8, name = 'Set2'))(length(unique(pd$country)))
# cols <- rainbow(3)
cols <- c( '#FF6633', 'darkred', '#006666')
ggplot(data = pd_big,
aes(x = days_since_start_deaths,
y = deaths,
group = ccaa)) +
geom_line(aes(color = country),
alpha = 0.9,
size = 2) +
geom_line(data = pd_small,
aes(x = days_since_start_deaths,
y = deaths,
color = country),
alpha = 0.7,
size = 1) +
scale_y_log10() +
scale_color_manual(name = '',
values = c(cols)) +
theme_simple() +
theme(legend.position = 'top') +
labs(x = 'Dias desde el primer día con 5 o más muertes acumuladas',
y = 'Muertes',
title = 'Muertes por COVID-19',
caption = '@joethebrew | www.databrew.cc') +
geom_text(data = label_data_big,
aes(x = days_since_start_deaths + 1.6,
y = ifelse(ccaa == 'Hubei', (deaths -500),
ifelse(ccaa == 'Lombardia', (deaths + 700),
(deaths + 300))),
label = gsub(' ', '\n', ccaa),
color = country),
size = 8,
alpha = 0.9,
show.legend = FALSE) +
geom_text(data = label_data_small,
aes(x = days_since_start_deaths + 1.6,
align = 'left',
y = deaths ,
label = ccaa,
# label = gsub(' ', '\n', ccaa),
color = country),
size = 5,
alpha = 0.6,
show.legend = FALSE) +
theme(axis.text = element_text(size = 14),
axis.title = element_text(size = 20),
legend.text = element_text(size = 16),
plot.title = element_text(size = 30)) +
xlim(0, 35)
ggsave('~/Desktop/china_spain_italy_comparison_raw.png',
height = 7,
width = 10)
# Spanish data
a <- esp_df %>%
left_join(esp_pop) %>%
mutate(country = 'Spain')
# Italian data
b <- ita %>%
left_join(ita_pop) %>%
mutate(country = 'Italy')
# Chinese data
d <- df %>% filter(country == 'China') %>%
mutate(cases = cases) %>%
mutate(ccaa = district) %>%
mutate(country = 'China') %>%
left_join(chi_pop)
# join
joined <- bind_rows(a, b, d)
# Get deaths per milllion
joined$deaths_pm <- joined$deaths / joined$pop * 1000000
joined$cases_pm <- joined$cases / joined$pop * 1000000
# Get the days since paradigm
x_deaths <- 5
x_deaths_pm <- 5
x_cases <- 50
x_cases_pm <- 50
joined <- joined %>%
arrange(date) %>%
group_by(ccaa) %>%
mutate(start_deaths = min(date[deaths >= x_deaths]),
start_cases = min(date[cases >= x_cases]),
start_deaths_pm = min(date[deaths_pm >= x_deaths_pm]),
start_cases_pm = min(date[cases_pm >= x_cases_pm])) %>%
ungroup %>%
mutate(days_since_start_deaths = date - start_deaths,
days_since_start_cases = date - start_cases,
days_since_start_deaths_pm = date - start_deaths_pm,
days_since_start_cases_pm = date - start_cases_pm)
# Define plot data
pd <- joined %>% filter(days_since_start_deaths >= 0) %>%
mutate(country = ifelse(country == 'China',
'China',
ifelse(country == 'Italy', 'Italia', 'España')))
add_zero <- function(x, n){
x <- as.character(x)
adders <- n - nchar(x)
adders <- ifelse(adders < 0, 0, adders)
for (i in 1:length(x)){
if(!is.na(x[i])){
x[i] <- paste0(
paste0(rep('0', adders[i]), collapse = ''),
x[i],
collapse = '')
}
}
return(x)
}
# # Define label data
# label_data <- pd %>% group_by(ccaa) %>% filter(
# (
# (country == 'China' & deaths >10 & days_since_start_deaths == 29) |
# (date == max(date) & country == 'España' & deaths > 90) |
# (date == max(date) & country == 'Italia' &
# ccaa != 'Liguria' & days_since_start_deaths > 10)
# ))
# # Get differential label data based on what to be emphasized
# label_data_big <- label_data %>%
# filter(ccaa %in% c('Madrid', 'Lombardia', 'Hubei'))
# label_data_small <- label_data %>%
# filter(!ccaa %in% c('Madrid', 'Lombardia', 'Hubei'))
#
pd_big <- pd %>%
filter(ccaa %in% c('Madrid', 'Lombardia', 'Hubei'))
pd_small <- pd %>%
filter(!ccaa %in% c('Madrid', 'Lombardia', 'Hubei'))
# cols <- colorRampPalette(RColorBrewer::brewer.pal(n = 8, name = 'Set2'))(length(unique(pd$country)))
# cols <- rainbow(3)
cols <- c( '#FF6633', 'darkred', '#006666')
the_dir <- '~/Desktop/animation/'
dir.create(the_dir)
the_dates <- sort(unique(c(pd_big$date, pd_small$date)))
for(i in 1:length(the_dates)){
the_date <- the_dates[i]
pd_big_sub <- pd_big %>% filter(date <= the_date)
pd_big_current <- pd_big_sub %>% filter(date == the_date)
pd_small_sub <- pd_small %>% filter(date <= the_date)
pd_small_current <- pd_small_sub %>% filter(date == the_date)
label_data_big <-
pd_big_sub %>%
filter(ccaa %in% c('Lombardia', 'Madrid', 'Hubei')) %>%
group_by(ccaa) %>%
filter(date == max(date)) %>%
ungroup %>%
mutate(days_since_start_deaths = ifelse(ccaa == 'Hubei' &
days_since_start_deaths >32,
32,
days_since_start_deaths))
label_data_small <-
pd_small_sub %>%
filter(ccaa %in% c('Emilia Romagna',
'Cataluña',
'CLM',
'País Vasco',
'Veneto',
'Piemonte',
'Henan',
'Heilongjiang')) %>%
group_by(ccaa) %>%
filter(date == max(date))
n_countries <- length(unique(pd_big_sub$country))
if(n_countries == 3){
sub_cols <- cols
}
if(n_countries == 2){
sub_cols <- cols[c(1,3)]
}
if(n_countries == 1){
sub_cols <- cols[1]
}
g <- ggplot(data = pd_big_sub,
aes(x = days_since_start_deaths,
y = deaths,
group = ccaa)) +
geom_line(aes(color = country),
alpha = 0.9,
size = 2) +
geom_line(data = pd_small_sub,
aes(x = days_since_start_deaths,
y = deaths,
color = country),
alpha = 0.7,
size = 1) +
geom_point(data = pd_big_current,
aes(x = days_since_start_deaths,
y = deaths,
color = country),
size = 3) +
geom_point(data = pd_small_current,
aes(x = days_since_start_deaths,
y = deaths,
color = country),
size = 1, alpha = 0.6) +
scale_y_log10(limits = c(5, 4500)) +
scale_color_manual(name = '',
values = sub_cols) +
theme_simple() +
theme(legend.position = 'top') +
labs(x = 'Dias desde el primer día con 5 o más muertes acumuladas',
y = 'Muertes',
title = format(the_date, '%d %b'),
subtitle = 'Muertes por COVID-19',
caption = '@joethebrew | www.databrew.cc') +
geom_text(data = label_data_big,
aes(x = days_since_start_deaths + 1,
y = deaths,
hjust = 0,
label = gsub(' ', '\n', ccaa),
color = country),
size = 8,
alpha = 0.9,
show.legend = FALSE) +
geom_text(data = label_data_small,
aes(x = days_since_start_deaths + 1.6,
y = deaths ,
label = ccaa,
# label = gsub(' ', '\n', ccaa),
color = country),
size = 5,
alpha = 0.6,
show.legend = FALSE) +
theme(axis.text = element_text(size = 14),
axis.title = element_text(size = 20),
legend.text = element_text(size = 16),
plot.title = element_text(size = 35),
plot.subtitle = element_text(size = 24)) +
xlim(0, 38)
message(i)
ggsave(paste0(the_dir, add_zero(i, 3), '.png'),
height = 7,
width = 10.5)
}
# Command line
cd ~/Desktop/animation
mogrify -resize 50% *.png
convert -delay 20 -loop 0 *.png result.gif
# Spanish data
a <- esp_df %>%
left_join(esp_pop) %>%
mutate(country = 'Spain')
joined <- a
# Get deaths per milllion
joined$deaths_pm <- joined$deaths / joined$pop * 1000000
joined$cases_pm <- joined$cases / joined$pop * 1000000
# Get the days since paradigm
x_deaths <- 5
x_deaths_pm <- 5
x_cases <- 50
x_cases_pm <- 50
joined <- joined %>%
arrange(date) %>%
group_by(ccaa) %>%
mutate(start_deaths = min(date[deaths >= x_deaths]),
start_cases = min(date[cases >= x_cases]),
start_deaths_pm = min(date[deaths_pm >= x_deaths_pm]),
start_cases_pm = min(date[cases_pm >= x_cases_pm])) %>%
ungroup %>%
mutate(days_since_start_deaths = date - start_deaths,
days_since_start_cases = date - start_cases,
days_since_start_deaths_pm = date - start_deaths_pm,
days_since_start_cases_pm = date - start_cases_pm)
# Define plot data
pd <- joined %>% filter(days_since_start_deaths >= 0) %>%
mutate(country = ifelse(country == 'China',
'China',
ifelse(country == 'Italy', 'Italia', 'España')))
bigs <- c('Madrid', 'Cataluña', 'CLM', 'CyL', 'País Vasco', 'La Rioja')
pd_big <- pd %>%
filter(ccaa %in% bigs)
pd_small <- pd %>%
filter(!ccaa %in% bigs)
# cols <- colorRampPalette(RColorBrewer::brewer.pal(n = 8, name = 'Set2'))(length(unique(pd$country)))
# cols <- rainbow(3)
cols <- colorRampPalette(c('#A16928','#bd925a','#d6bd8d','#edeac2', '#b5c8b8','#79a7ac','#2887a1'))(length(unique(pd$country)))
the_dir <- '~/Desktop/animation2/'
dir.create(the_dir)
the_dates <- sort(unique(c(pd_big$date, pd_small$date)))
for(i in 1:length(the_dates)){
the_date <- the_dates[i]
pd_big_sub <- pd_big %>% filter(date <= the_date)
pd_big_current <- pd_big_sub %>% filter(date == the_date)
pd_small_sub <- pd_small %>% filter(date <= the_date)
pd_small_current <- pd_small_sub %>% filter(date == the_date)
label_data_big <-
pd_big_sub %>%
filter(ccaa %in% bigs) %>%
group_by(ccaa) %>%
filter(date == max(date)) %>%
ungroup
label_data_small <-
pd_small_sub %>%
group_by(ccaa) %>%
filter(date == max(date))
# sub_cols <- colorRampPalette(c('#A16928','#bd925a','#d6bd8d','#edeac2', '#b5c8b8','#79a7ac','#2887a1'))(length(unique(pd$ccaa)))
sub_cols <- colorRampPalette(RColorBrewer::brewer.pal(n = 8, name = 'Dark2'))(length(unique(pd$ccaa)))
# sub_cols <- rainbow((length(unique(pd$ccaa))))
g <- ggplot(data = pd_big_sub,
aes(x = days_since_start_deaths,
y = deaths,
group = ccaa)) +
geom_line(aes(color = ccaa),
alpha = 0.9,
size = 2) +
geom_line(data = pd_small_sub,
aes(x = days_since_start_deaths,
y = deaths,
color = ccaa),
alpha = 0.7,
size = 1) +
geom_point(data = pd_big_current,
aes(x = days_since_start_deaths,
y = deaths,
color = ccaa),
size = 3) +
geom_point(data = pd_small_current,
aes(x = days_since_start_deaths,
y = deaths,
color = ccaa),
size = 1, alpha = 0.6) +
geom_point(data = pd,
aes(x = days_since_start_deaths,
y = deaths,
color = ccaa),
size = 1, alpha = 0.01) +
scale_y_log10(limits = c(5, max(pd$deaths)*1.2),
breaks = c(10, 50, 100, 500, 1000)) +
scale_color_manual(name = '',
values = sub_cols) +
theme_simple() +
theme(legend.position = 'top') +
labs(x = 'Dias desde el primer día con 5 o más muertes acumuladas',
y = 'Muertes',
title = format(the_date, '%d %b'),
subtitle = 'Muertes por COVID-19',
caption = '@joethebrew | www.databrew.cc') +
theme(axis.text = element_text(size = 14),
axis.title = element_text(size = 20),
legend.text = element_text(size = 16),
plot.title = element_text(size = 35),
plot.subtitle = element_text(size = 24)) +
xlim(0, 20) +
theme(legend.position = 'none')
message(i)
if(nrow(label_data_big) > 0){
g <- g +
geom_text(data = label_data_big,
aes(x = days_since_start_deaths + 0.2,
y = deaths,
hjust = 0,
label = gsub(' ', ' ', ccaa),
color = ccaa),
size = 8,
alpha = 0.9,
show.legend = FALSE) +
geom_text(data = label_data_small,
aes(x = days_since_start_deaths + 0.2,
y = deaths ,
label = ccaa,
# label = gsub(' ', '\n', ccaa),
color = ccaa),
size = 5,
alpha = 0.6,
show.legend = FALSE)
}
ggsave(paste0(the_dir, add_zero(i, 3), '.png'),
height = 7,
width = 12)
}
# Command line
cd ~/Desktop/animation
mogrify -resize 50% *.png
convert -delay 25 -loop 0 *.png result.gif
# Spanish data
a <- esp_df %>%
left_join(esp_pop) %>%
mutate(country = 'Spain')
# Italian data
b <- ita %>%
left_join(ita_pop) %>%
mutate(country = 'Italy')
# join
joined <- bind_rows(a, b)
# Get deaths per milllion
joined$deaths_pm <- joined$deaths / joined$pop * 1000000
joined$cases_pm <- joined$cases / joined$pop * 1000000
# Get the days since paradigm
x_deaths <- 5
x_deaths_pm <- 5
x_cases <- 50
x_cases_pm <- 50
joined <- joined %>%
arrange(date) %>%
group_by(ccaa) %>%
mutate(start_deaths = min(date[deaths >= x_deaths]),
start_cases = min(date[cases >= x_cases]),
start_deaths_pm = min(date[deaths_pm >= x_deaths_pm]),
start_cases_pm = min(date[cases_pm >= x_cases_pm])) %>%
ungroup %>%
mutate(days_since_start_deaths = date - start_deaths,
days_since_start_cases = date - start_cases,
days_since_start_deaths_pm = date - start_deaths_pm,
days_since_start_cases_pm = date - start_cases_pm)
ggplot(data = joined %>% filter(days_since_start_deaths_pm >= 0),
aes(x = days_since_start_deaths_pm,
y = deaths_pm,
group = ccaa)) +
geom_line(aes(color = country),
alpha = 0.8,
size = 2) +
scale_y_log10() +
scale_color_manual(name = '',
values = c('darkorange', 'purple')) +
theme_simple() +
theme(legend.position = 'none') +
labs(x = 'Days since "start out outbreak"',
y = 'Deaths per million',
title = 'Deaths per capita, Italian regions vs. Spanish autonomous communities',
subtitle = paste0('Day 0 ("start of outbreak") = first day at ', x_deaths_pm, ' or greater cumulative deaths per million'),
caption = '@joethebrew | www.databrew.cc') +
geom_text(data = joined %>% group_by(ccaa) %>% filter(date == max(date) &
(
(country == 'Spain' & deaths_pm > 25) |
(country == 'Italy' & days_since_start_deaths_pm > 10)
)),
aes(x = days_since_start_deaths_pm + 0.6,
y = deaths_pm,
label = gsub(' ', '\n', ccaa),
color = country),
size = 6) +
theme(axis.text = element_text(size = 14),
axis.title = element_text(size = 20)) +
xlim(0, 23)
ggsave('~/Desktop/italy_comparison.png',
height = 6,
width = 10)
# Separate for Catalonia
pd <- joined %>% filter(days_since_start_deaths_pm >= 0) %>%
mutate(country = ifelse(ccaa == 'Cataluña',
'Catalonia',
country)) %>%
mutate(ccaa = ifelse(ccaa == 'Cataluña', 'Catalunya', ccaa))
pdcat <- pd %>% filter(country == 'Catalonia')
label_data <- pd %>% group_by(ccaa) %>% filter(date == max(date) &
(
(country == 'Catalonia') |
(country == 'Spain' & deaths_pm > 25) |
(country == 'Italy' & days_since_start_deaths_pm > 10)
))
ggplot(data = pd,
aes(x = days_since_start_deaths_pm,
y = deaths_pm,
group = ccaa)) +
geom_line(aes(color = country),
alpha = 0.3,
size = 1.5) +
geom_line(data = pdcat,
aes(color = country),
alpha = 0.8,
size = 2) +
geom_point(data = pdcat %>% filter(date == max(date)),
aes(color = country),
alpha = 0.8,
size = 4) +
scale_y_log10() +
scale_color_manual(name = '',
values = c('darkred', 'darkorange', "purple")) +
theme_simple() +
theme(legend.position = 'none') +
labs(x = 'Dies des del "començament del brot"',
y = 'Morts per milió',
title = 'Morts per càpita: Catalunya, comunitats autònomes, regions italianes',
subtitle = paste0('Dia 0 ("començament del brot") = primer dia a ', x_deaths_pm, ' o més morts acumulades per milió de població'),
caption = '@joethebrew | www.databrew.cc') +
geom_text(data = label_data,
aes(x = days_since_start_deaths_pm +0.2 ,
y = deaths_pm +3,
hjust = 0,
label = gsub(' ', '\n', ccaa),
color = country),
size = 6,
alpha = 0.7) +
theme(axis.text = element_text(size = 14),
axis.title = element_text(size = 20)) +
xlim(0, 24)
ggsave('~/Desktop/cat_italy_comparison.png',
height = 6,
width = 10)
# Straightforward Lombardy, Madrid, Cat comparison
specials <- c('Lombardia', 'Madrid')
pd <- joined %>% filter(days_since_start_deaths_pm >= 0) %>%
mutate(country = ifelse(ccaa == 'Cataluña',
'Catalonia',
country)) %>%
mutate(ccaa = ifelse(ccaa == 'Cataluña', 'Catalunya', ccaa))
pdcat <- pd %>% filter(ccaa %in% specials)
label_data <- pd %>% group_by(ccaa) %>% filter(date == max(date) &
(
# (country == 'Catalonia') |
(country == 'Spain' & deaths_pm > 20) |
(country == 'Italy' & days_since_start_deaths_pm >= 10)
))
ggplot(data = pd,
aes(x = days_since_start_deaths_pm,
y = deaths_pm,
group = ccaa)) +
geom_line(aes(color = country),
alpha = 0.3,
size = 1.5) +
geom_line(data = pdcat,
aes(color = country),
alpha = 0.8,
size = 2) +
scale_y_log10() +
scale_color_manual(name = '',
values = c('darkred', 'darkorange', "purple")) +
theme_simple() +
theme(legend.position = 'none') +
labs(x = 'Dias desde "el comienzo del brote"',
y = 'Muertes por millón de habitantes',
title = 'Muertes acumuladas por 1.000.000 habitantes',
subtitle = paste0('Dia 0 ("comienzo del brote") = primer día a ', x_deaths_pm, ' o más muertes acumuladas por milión de población'),
caption = '@joethebrew | www.databrew.cc') +
geom_text(data = label_data %>% filter(!ccaa %in% specials),
aes(x = days_since_start_deaths_pm + 0.4,
y = deaths_pm +3,
label = gsub(' ', '\n', ccaa),
color = country),
size = 5,
alpha = 0.5) +
geom_text(data = label_data %>% filter(ccaa %in% specials),
aes(x = days_since_start_deaths_pm ,
y = deaths_pm +30,
label = gsub(' ', '\n', ccaa),
color = country),
size = 8,
alpha = 0.8) +
theme(axis.text = element_text(size = 14),
axis.title = element_text(size = 20)) +
xlim(0, 23)
ggsave('~/Desktop/mad_lom_italy_comparison.png',
height = 6,
width = 10)
isos <- sort(unique(world_pop$sub_region))
isos <- c('Central Asia', 'Eastern Asia', 'Eastern Europe',
'Latin America and the Caribbean',
'Northern Africa', 'Northern America',
'Nothern Europe',
'South-eastern Asia',
'Southern Asia', 'Southern Europe',
'Sub-Saharan Africa', 'Western Asia', 'Western Europe')
dir.create('~/Desktop/world')
for(i in 1:length(isos)){
this_iso <- isos[i]
message(i, ' ', this_iso)
countries <- world_pop %>% filter(sub_region == this_iso)
pd <- df %>%
group_by(country, iso, date) %>%
summarise(cases = sum(cases, na.rm = TRUE)) %>%
ungroup %>%
group_by(country) %>%
filter(length(which(cases > 0)) > 1) %>%
ungroup %>%
filter(iso %in% countries$iso)
if(nrow(pd) > 0){
cols <- colorRampPalette(brewer.pal(n = 8, 'Spectral'))(length(unique(pd$country)))
cols <- sample(cols, length(cols))
# Plot
n_countries <- (length(unique(pd$country)))
ggplot(data = pd,
aes(x = date,
# color = country,
# fill = country,
y = cases)) +
theme_simple() +
# geom_point() +
# geom_line() +
geom_area(fill = 'darkred', alpha = 0.3, color = 'darkred') +
# scale_color_manual(name = '',
# values = cols) +
# scale_fill_manual(name = '',
# values = cols) +
theme(legend.position = 'none',
axis.text = element_text(size = 6),
strip.text = element_text(size = ifelse(n_countries > 20, 6,
ifelse(n_countries > 10, 10,
ifelse(n_countries > 5, 11, 12))) ),
legend.text = element_text(size = 6)) +
# scale_y_log10() +
facet_wrap(~country,
scales = 'free') +
labs(x = '',
y = 'Confirmed cases',
title = paste0('Confirmed cases of COVID-19 in ', this_iso))
ggsave(paste0('~/Desktop/world/', this_iso, '.png'),
width = 12,
height = 7)
}
}
roll_curve <- function(data,
n = 4,
deaths = FALSE,
scales = 'fixed',
nrow = NULL,
ncol = NULL,
pop = FALSE){
# Create the n day rolling avg
ma <- function(x, n = 5){
if(length(x) >= n){
stats::filter(x, rep(1 / n, n), sides = 1)
} else {
new_n <- length(x)
stats::filter(x, rep(1 / new_n, new_n), sides = 1)
}
}
pd <- data
if(deaths){
pd$var <- pd$deaths_non_cum
} else {
pd$var <- pd$cases_non_cum
}
if('ccaa' %in% names(pd)){
pd$geo <- pd$ccaa
} else {
pd$geo <- pd$iso
}
if(pop){
# try to get population
if(any(pd$geo %in% unique(esp_df$ccaa))){
right <- esp_pop
right_var <- 'ccaa'
} else {
right <- world_pop
right_var <- 'iso'
}
pd <- pd %>% left_join(right %>% dplyr::select(all_of(right_var), pop),
by = c('geo' = right_var)) %>%
mutate(var = var / pop * 100000)
}
pd <- pd %>%
arrange(date) %>%
group_by(geo) %>%
mutate(with_lag = ma(var, n = n))
ggplot() +
geom_bar(data = pd,
aes(x = date,
y = var),
stat = 'identity',
fill = 'grey',
alpha = 0.8) +
geom_area(data = pd,
aes(x = date,
y = with_lag),
color = 'red',
fill = 'red',
alpha = 0.6) +
facet_wrap(~geo, scales = scales, nrow = nrow, ncol = ncol) +
theme_simple() +
labs(x = '',
y = ifelse(deaths, 'Deaths', 'Cases'),
title = paste0('Daily (non-cumulative) ', ifelse(deaths, 'deaths', 'cases'),
ifelse(pop, ' (per 100,000)', '')),
subtitle = paste0('Data as of ', max(data$date),
'.\nRed line = ', n, ' day rolling average. Grey bar = day-specific value.')) +
theme(axis.text.x = element_text(size = 7,
angle = 90,
hjust = 0.5, vjust = 1)) #+
# scale_x_date(name ='',
# breaks = sort(unique(pd$date)),
# labels = format(sort(unique(pd$date)), '%b %d'))
# scale_y_log10()
}
plot_data <- df_country %>% filter(country %in% c('Spain', 'France', 'Italy', 'Germany', 'Belgium', 'Norway')) %>% mutate(geo = country)
roll_curve(plot_data, pop = T)
dir.create('~/Desktop/countries')
roll_curve_country <- function(the_country = 'Spain'){
plot_data <- df_country %>% filter(country %in% the_country) %>% mutate(geo = country)
g1 <- roll_curve(plot_data, pop = F)
g2 <- roll_curve(plot_data, pop = T)
g3 <- roll_curve(plot_data, pop = F, deaths = T)
g4 <- roll_curve(plot_data, pop = T, deaths = T)
ggsave(paste0('~/Desktop/countries/', the_country, '1.png'), g1)
ggsave(paste0('~/Desktop/countries/', the_country, '2.png'), g2)
ggsave(paste0('~/Desktop/countries/', the_country, '3.png'), g3)
ggsave(paste0('~/Desktop/countries/', the_country, '4.png'), g4)
}
countries <- c('Spain', 'France', 'Italy', 'Germany', 'Belgium', 'Norway', 'US', 'United Kingdom', 'Korea, South',
'China', 'Japan', 'Switzerland', 'Sweden', 'Denmark', 'Netherlands', 'Iran', 'Canada')
for(i in 1:length(countries)){
roll_curve_country(the_country = countries[i])
}
Error in ts(x): 'ts' object must have one or more observations
# Cases
plot_data <- df_country %>% filter(country == 'Spain') %>% mutate(geo = country)
roll_curve(plot_data)
# Cases adjusted
plot_data <- df_country %>% filter(country == 'Spain') %>% mutate(geo = country)
roll_curve(plot_data, pop = T)
# Deaths
plot_data <- df_country %>% filter(country == 'Spain') %>% mutate(geo = country)
roll_curve(plot_data, deaths = T)
# Cases adjusted
plot_data <- df_country %>% filter(country == 'Spain') %>% mutate(geo = country)
roll_curve(plot_data, pop = T, deaths = T)
plot_data <- esp_df %>% mutate(geo = ccaa)
roll_curve(plot_data, pop = T, deaths = T)
plot_data <- df_country %>% filter(country == 'Spain') %>% mutate(geo = country)
roll_curve(plot_data, deaths = T)
# Latest in Spain
pd <- esp_df %>%
filter(date == max(date)) %>%
mutate(p = deaths / sum(deaths) * 100)
text_size <- 12
# deaths
ggplot(data = pd,
aes(x = ccaa,
y = deaths)) +
geom_bar(stat = 'identity',
fill = 'black') +
theme_simple() +
theme(axis.text.x = element_text(angle = 90, hjust = 1, vjust = 0.5)) +
labs(x = '',
y = 'Deaths | Muertes',
title = 'COVID-19 deaths in Spain',
subtitle = paste0('Data as of ', max(pd$date)),
caption = 'github.com/databrew/covid19 | joe@databrew.cc') +
theme(legend.position = 'top',
legend.text = element_text(size = text_size * 2),
axis.title = element_text(size = text_size * 2),
plot.title = element_text(size = text_size * 2.3),
axis.text.x = element_text(size = text_size * 1.5)) +
geom_text(data = pd %>% filter(deaths > 0),
aes(x = ccaa,
y = deaths,
label = paste0(deaths, '\n(',
round(p, digits = 1), '%)')),
size = text_size * 0.3,
nudge_y = 180) +
ylim(0, 2500)
ggsave('~/Desktop/spain.png')
Muertes relativas por CCAA
# Latest in Spain
pd <- esp_df %>%
filter(date == max(date)) %>%
mutate(p = deaths / sum(deaths) * 100)
pd <- pd %>% left_join(esp_pop)
text_size <- 12
pd$value <- pd$deaths / pd$pop * 100000
# deaths
ggplot(data = pd,
aes(x = ccaa,
y = value)) +
geom_bar(stat = 'identity',
fill = 'black') +
theme_simple() +
theme(axis.text.x = element_text(angle = 90, hjust = 1, vjust = 0.5)) +
labs(x = '',
y = 'Deaths per 100,000',
title = 'COVID-19 deaths per 100.000',
subtitle = paste0('Data as of ', max(pd$date)),
caption = 'github.com/databrew/covid19 | joe@databrew.cc') +
theme(legend.position = 'top',
legend.text = element_text(size = text_size * 2),
axis.title = element_text(size = text_size * 2),
plot.title = element_text(size = text_size * 2.3),
axis.text.x = element_text(size = text_size * 1.5)) +
geom_text(data = pd %>% filter(value > 0),
aes(x = ccaa,
y = value,
label = paste0(round(value, digits = 2), '\n(',
deaths, '\ndeaths)')),
size = text_size * 0.3,
nudge_y = 4.5) +
ylim(0, 40)
ggsave('~/Desktop/spai2.png')
plot_data <- esp_df %>% mutate(geo = ccaa) %>% filter(!ccaa %in% c('Melilla'))
roll_curve(plot_data, scales = 'fixed')
ggsave('~/Desktop/a.png',
width = 1280 / 150,
height = 720 / 150)
Loop for everywhere (see desktop)
dir.create('~/Desktop/ccaas')
ccaas <- sort(unique(esp_df$ccaa))
for(i in 1:length(ccaas)){
this_ccaa <- ccaas[i]
plot_data <- esp_df %>% mutate(geo = ccaa) %>% filter(ccaa == this_ccaa)
roll_curve(plot_data, scales = 'fixed') + theme(strip.text = element_text(size = 30))
ggsave(paste0('~/Desktop/ccaas/', i, this_ccaa, '_cases.png'),
width = 1280 / 150,
height = 720 / 150)
}
ccaas <- sort(unique(esp_df$ccaa))
for(i in 1:length(ccaas)){
this_ccaa <- ccaas[i]
plot_data <- esp_df %>% mutate(geo = ccaa) %>% filter(ccaa == this_ccaa)
roll_curve(plot_data, scales = 'fixed', pop = TRUE) + theme(strip.text = element_text(size = 30))
ggsave(paste0('~/Desktop/ccaas/', i, this_ccaa, '_cases_pop.png'),
width = 1280 / 150,
height = 720 / 150)
}
# Deaths too
for(i in 1:length(ccaas)){
this_ccaa <- ccaas[i]
plot_data <- esp_df %>% mutate(geo = ccaa) %>% filter(ccaa == this_ccaa)
roll_curve(plot_data, deaths = T, scales = 'fixed') + theme(strip.text = element_text(size = 30))
ggsave(paste0('~/Desktop/ccaas/', i, this_ccaa, '_deaths.png'),
width = 1280 / 150,
height = 720 / 150)
}
# Deaths too
for(i in 1:length(ccaas)){
this_ccaa <- ccaas[i]
plot_data <- esp_df %>% mutate(geo = ccaa) %>% filter(ccaa == this_ccaa)
roll_curve(plot_data, deaths = T, scales = 'fixed', pop = TRUE) + theme(strip.text = element_text(size = 30))
ggsave(paste0('~/Desktop/ccaas/', i, this_ccaa, '_deaths_pop.png'),
width = 1280 / 150,
height = 720 / 150)
}
plot_data <- esp_df %>% mutate(geo = ccaa) %>% filter(!ccaa %in% c('Melilla'))
roll_curve(plot_data, scales = 'free_y')
ggsave('~/Desktop/b.png',
width = 1280 / 150,
height = 720 / 150)
plot_data <- esp_df %>% mutate(geo = ccaa) %>% filter(!ccaa %in% c('Melilla'))
roll_curve(plot_data, deaths = T, scales = 'free_y')
ggsave('~/Desktop/c.png',
width = 1280 / 150,
height = 720 / 150)
plot_data <- esp_df %>% mutate(geo = ccaa) %>% filter(!ccaa %in% c('Melilla'))
roll_curve(plot_data, deaths = T, scales = 'fixed')
ggsave('~/Desktop/d.png',
width = 1280 / 150,
height = 720 / 150)
plot_data <- df_country %>% filter(country %in% c('Spain', 'Italy', 'Germany', 'France', 'US',
'China', 'Korea, South', 'Japan', 'Singapore')) %>% mutate(geo = country)
roll_curve(plot_data, scales = 'free_y')
ggsave('~/Desktop/z.png',
width = 1280 / 150,
height = 720 / 150)
pd <- df_country %>%
group_by(date) %>%
summarise(n = sum(cases)) %>%
filter(date < max(date))
ggplot(data = pd,
aes(x = date,
y = n)) +
geom_point() +
theme_simple() +
labs(x = 'Date',
y = 'Cases',
title = 'COVID-19 cases')
ggsave('~/Videos/update/a.png',
width = 1280 / 150,
height = 720 / 150)
Error in grid.newpage(): could not open file '/home/joebrew/Videos/update/a.png'
pd <- df_country %>%
group_by(date,
country = ifelse(country == 'China', 'China', 'Other countries')) %>%
summarise(n = sum(cases)) %>%
ungroup %>%
filter(date < max(date))
Error: Column `country` can't be modified because it's a grouping variable
ggplot(data = pd,
aes(x = date,
y = n,
color = country)) +
geom_line(size = 2) +
# geom_point() +
theme_simple() +
labs(x = 'Date',
y = 'Cases',
title = 'COVID-19 cases') +
scale_color_manual(name = '',
values = c('red', 'black')) +
theme(legend.text = element_text(size = 25),
legend.position = 'top')
Error in FUN(X[[i]], ...): object 'country' not found
ggsave('~/Videos/update/b.png',
width = 1280 / 150,
height = 720 / 150)
Error in grid.newpage(): could not open file '/home/joebrew/Videos/update/b.png'
pd <- df_country %>%
group_by(date,
country = ifelse(country == 'China', 'China', 'Other countries')) %>%
summarise(n = sum(cases)) %>%
filter(country == 'Other countries') %>%
ungroup %>%
filter(date < max(date))
Error: Column `country` can't be modified because it's a grouping variable
ggplot(data = pd,
aes(x = date,
y = n)) +
geom_line(size = 2) +
# geom_point() +
theme_simple() +
labs(x = 'Date',
y = 'Cases',
title = 'COVID-19 cases, outside of China')
ggsave('~/Videos/update/c.png',
width = 1280 / 150,
height = 720 / 150)
Error in grid.newpage(): could not open file '/home/joebrew/Videos/update/c.png'
plot_day_zero(countries = c('France', 'Germany', 'Italy', 'Spain', 'Switzerland', 'Sweden', 'Norway', 'Netherlands'))
# ggsave('~/Videos/update/d.png',
# width = 1280 / 150,
# height = 720 / 150)
pd <- df_country %>%
group_by(date) %>%
summarise(n = sum(deaths)) %>%
filter(date < max(date))
ggplot(data = pd,
aes(x = date,
y = n)) +
geom_point() +
theme_simple() +
labs(x = 'Date',
y = 'Deaths',
title = 'COVID-19 deaths')
# ggsave('~/Videos/update/e.png',
# width = 1280 / 150,
# height = 720 / 150)
pd <- df_country %>%
group_by(date,
country = ifelse(country == 'China', 'China', 'Other countries')) %>%
summarise(n = sum(deaths)) %>%
ungroup %>%
filter(date < max(date))
Error: Column `country` can't be modified because it's a grouping variable
ggplot(data = pd,
aes(x = date,
y = n,
color = country)) +
geom_line(size = 2) +
# geom_point() +
theme_simple() +
labs(x = 'Date',
y = 'Deaths',
title = 'COVID-19 deaths') +
scale_color_manual(name = '',
values = c('red', 'black')) +
theme(legend.text = element_text(size = 25),
legend.position = 'top')
Error in FUN(X[[i]], ...): object 'country' not found
# ggsave('~/Videos/update/f.png',
# width = 1280 / 150,
# height = 720 / 150)
plot_day_zero(countries = c('Korea, South', 'Japan', 'China', 'Singapore'))
# ggsave('~/Videos/update/g.png',
# width = 1280 / 150,
# height = 720 / 150)
Since trajectories are very unstable when cases are low, we’ll exclude from our analysis the first few days, and will only count as “outbreak” once a country reaches 150 or more cumulative cases.
# Doubling time
n_cases_start = 150
countries = c('Italy', 'Spain', 'France', 'Germany', 'Italy', 'Switzerland', 'Denmark', 'US', 'United Kingdom', 'Norway')
# countries <- sort(unique(df_country$country))
out_list <- curve_list <- list()
counter <- 0
for(i in 1:length(countries)){
message(i)
this_country <- countries[i]
sub_data <-df_country %>% filter(country == this_country)
# Only calculate on countries with n_cases_start or greater cases,
# starting at the first day at n_cases_start or greater
ok <- max(sub_data$cases, na.rm = TRUE) >= n_cases_start
if(ok){
counter <- counter + 1
pd <- sub_data %>%
filter(!is.na(cases)) %>%
mutate(start_date = min(date[cases >= n_cases_start])) %>%
mutate(days_since = date - start_date) %>%
filter(days_since >= 0) %>%
mutate(days_since = as.numeric(days_since))
fit <- lm(log(cases) ~ days_since, data = pd)
# plot(pd$days_since, log(pd$cases))
# abline(fit)
## Slope
curve <- fit$coef[2]
# Predict days ahead
fake <- tibble(days_since = seq(0, max(pd$days_since) + 5, by = 1))
fake <- left_join(fake, pd %>% dplyr::select(days_since, cases, date))
fake$predicted <- exp(predict(fit, newdata = fake))
# Doubling time
dt <- log(2)/fit$coef[2]
out <- tibble(country = this_country,
doubling_time = dt,
slope = curve)
out_list[[counter]] <- out
curve_list[[counter]] <- fake %>% mutate(country = this_country)
}
}
done <- bind_rows(out_list)
print(done)
# A tibble: 10 x 3
country doubling_time slope
<chr> <dbl> <dbl>
1 Italy 3.65 0.190
2 Spain 2.51 0.276
3 France 3.27 0.212
4 Germany 2.91 0.238
5 Italy 3.65 0.190
6 Switzerland 3.24 0.214
7 Denmark 6.96 0.0995
8 US 2.32 0.298
9 United Kingdom 3.16 0.219
10 Norway 4.80 0.144
curves <- bind_rows(curve_list)
# Get curves back in exponential form
# curves$curve <- exp(curves$curve)
# Join doubling time to curves
joined <- left_join(curves, done)
# Get rid of Italy future (since it's the "leader")
joined <- joined %>%
filter(country != 'Italy' |
date <= (Sys.Date() -1))
# Make long format
long <- joined %>%
dplyr::select(date, days_since, country, cases, predicted, doubling_time) %>%
tidyr::gather(key, value, cases:predicted) %>%
mutate(key = Hmisc::capitalize(gsub('_', ' ', key))) %>%
mutate(key = ifelse(key == 'Predicted', 'Predicted (based on current doubling time)', key))
The below chart shows the trajectories in terms of number of cases in Europe in red, and the predicted trajectories in black. The black line assumes that the doubling rate will stay constant.
cols <- c('red', 'black')
ggplot(data = long,
aes(x = days_since,
y = value,
lty = key,
color = key)) +
geom_line(data = long %>% filter(key != 'Confirmed cases'),
size = 1.2, alpha = 0.8) +
geom_point(data = long %>% filter(key == 'Confirmed cases')) +
geom_line(data = long %>% filter(key == 'Confirmed cases'),
size = 0.8) +
facet_wrap(~paste0(country, '\n',
'(doubling time: ',
round(doubling_time, digits = 1), ' days)'), scales = 'free') +
theme_simple() +
scale_linetype_manual(name ='',
values = c(1,2)) +
scale_color_manual(name = '',
values = cols) +
theme(legend.position = 'top') +
labs(x = 'Days since first day at >150 cumulative cases',
y = 'Cases',
title = 'COVID-19 CASES: ("predicted" assumes no change in doubling time)',
caption = 'Data from Johns Hopkins. Processing: Joe Brew @joethebrew. Code: github.com/databrew/covid19',
subtitle = '(Doubling time calculated since first day at >150 cumulative cases)') +
theme(strip.text = element_text(size = 13),
plot.title = element_text(size = 15))
Since Italy is “leading the way”, it’s helpful to also compare each country to Italy. Let’s see that.
# Overlay Italy
ol1 <- joined %>% filter(!country %in% 'Italy')
ol2 <- joined %>% filter(country == 'Italy') %>% dplyr::rename(Italy = cases) %>%
dplyr::select(Italy, days_since)
ol <- left_join(ol1, ol2) %>%
dplyr::select(days_since, date, country, cases, predicted, Italy,doubling_time)
ol <- tidyr::gather(ol, key, value, cases: Italy) %>%
mutate(key = Hmisc::capitalize(gsub('_', ' ', key))) %>%
mutate(key = ifelse(key == 'Predicted', 'Predicted (based on current doubling time)', key))
cols <- c('red', 'blue', 'black')
ggplot(data = ol,
aes(x = days_since,
y = value,
lty = key,
color = key)) +
geom_line(data = ol %>% filter(!key %in% c('Confirmed cases', 'Italy')),
size = 1.2, alpha = 0.8) +
geom_line(data = ol %>% filter(key %in% c('Italy')),
size = 0.8, alpha = 0.8) +
geom_point(data = ol %>% filter(key == 'Confirmed cases')) +
geom_line(data = ol %>% filter(key == 'Confirmed cases'),
size = 0.8) +
facet_wrap(~paste0(country, '\n',
'(doubling time: ',
round(doubling_time, digits = 1), ' days)'), scales = 'free') +
theme_simple() +
scale_linetype_manual(name ='',
values = c(1,6,2)) +
scale_color_manual(name = '',
values = cols) +
theme(legend.position = 'top') +
labs(x = 'Days since first day at >150 cumulative cases',
y = 'Cases',
title = 'COVID-19 CASES: ("predicted" assumes no change in doubling time)',
caption = 'Data from Johns Hopkins. Processing: Joe Brew @joethebrew. Code: github.com/databrew/covid19',
subtitle = '(Doubling time calculated since first day at >150 cumulative cases)') +
theme(strip.text = element_text(size = 13),
plot.title = element_text(size = 15))
In the above, what’s striking is how many places have trajectories that are worse than Italy’s. Yes, Italy has more cases, but it’s doubling time is less. Either that changes soon, or these other countries will soon have more cases than Italy.
The number of cases is not necessarily the best indicator for the severity of an outbreak of this nature. Why? Because (a) testing rates and protocols are different by place and (b) testing rates are different by time (since health services are changing their approaches as things develop). In other words, when we compare the number of cases by place and time, we are introducing significant bias.
Using deaths to gauge the magnitude of the outbreak is also problematic. Death rates are differential by age, so the number of deaths depends on a country’s population period, or age structure. Also, death rates will be a function of health services, which are not of the same quality every where. And, of course, like cases, we don’t necessarily know about all of the deaths that occur because of COVID-19.
Still, there’s an argument that death rates have less bias than case rates because deaths are easier to identify than cases. Let’s accept that argument, for the time being, and have a look at death rates by country.
# Doubling time
n_deaths_start = 5
countries = c('Italy', 'Spain', 'France', 'Italy', 'Switzerland', 'Denmark', 'US', 'United Kingdom', 'Norway', 'Germany')
# countries <- sort(unique(df_country$country))
make_double_time <- function(data = df_country,
the_country = 'Spain',
n_deaths_start = 5,
time_ahead = 7){
sub_data <-data %>% filter(country == the_country)
# Only calculate on countries with n_cases_start or greater cases,
# starting at the first day at n_cases_start or greater
ok <- max(sub_data$deaths, na.rm = TRUE) >= n_deaths_start
if(ok){
counter <- counter + 1
pd <- sub_data %>%
filter(!is.na(deaths)) %>%
mutate(start_date = min(date[deaths >= n_deaths_start])) %>%
mutate(days_since = date - start_date) %>%
filter(days_since >= 0) %>%
mutate(days_since = as.numeric(days_since)) %>%
mutate(the_weight = 1/(1 + (as.numeric(max(date) - date))))
fit <- lm(log(deaths) ~ days_since,
weights = the_weight,
data = pd)
# fitlo <- loess(deaths ~ days_since, data = pd)
# plot(pd$days_since, log(pd$cases))
# abline(fit)
## Slope
# curve <- fit$coef[2]
# Predict days ahead
day0 <- pd$date[pd$days_since == 0]
fake <- tibble(days_since = seq(0, max(pd$days_since) + time_ahead, by = 1))
fake <- fake %>%mutate(date = seq(day0, day0+max(fake$days_since), by = 1))
fake <- left_join(fake, pd %>% dplyr::select(days_since, deaths, date))
fake$predicted <- exp(predict(fit, newdata = fake))
# fake$predictedlo <- predict(fitlo, newdata = fake)
ci <- exp(predict(fit, newdata = fake, interval = 'prediction'))
# cilo <- predict(fitlo, newdata = fake, interval = 'prediction')
fake$lwr <- ci[,'lwr']
fake$upr <- ci[,'upr']
# fake$lwrlo <- ci[,'lwr']
# fake$uprlo <- ci[,'upr']
# Doubling time
dt <- log(2)/fit$coef[2]
fake %>% mutate(country = the_country) %>% mutate(doubling_time = dt)
}
}
plot_double_time <- function(data, ylog = F){
the_labs <- labs(x = 'Date',
y = 'Deaths',
title = paste0('Predicted deaths in ', data$country[1]))
long <- data %>%
tidyr::gather(key, value, deaths:predicted) %>%
mutate(key = Hmisc::capitalize(key))
g <- ggplot() +
geom_ribbon(data = data %>% filter(date > max(long$date[!is.na(long$value) & long$key == 'Deaths'])),
aes(x = date,
ymax = upr,
ymin = lwr),
alpha =0.6,
fill = 'darkorange') +
geom_line(data = long,
aes(x = date,
y = value,
group = key,
lty = key)) +
geom_point(data = long %>% filter(key == 'Deaths'),
aes(x = date,
y = value)) +
theme_simple() +
theme(legend.position = 'right',
legend.title = element_blank()) +
the_labs
if(ylog){
g <- g + scale_y_log10()
}
return(g)
}
options(scipen = '999')
data <- make_double_time(n_deaths_start = 150, time_ahead = 7)
data
# A tibble: 19 x 8
days_since date country deaths predicted lwr upr doubling_time
<dbl> <date> <chr> <dbl> <dbl> <dbl> <dbl> <dbl>
1 0 2020-03-15 Spain 288 385. 326. 453. 2.94
2 1 2020-03-16 Spain 491 487. 418. 567. 2.94
3 2 2020-03-17 Spain 598 616. 535. 710. 2.94
4 3 2020-03-18 Spain 767 780. 684. 889. 2.94
5 4 2020-03-19 Spain 1002 987. 874. 1115. 2.94
6 5 2020-03-20 Spain 1326 1249. 1115. 1400. 2.94
7 6 2020-03-21 Spain 1720 1581. 1420. 1761. 2.94
8 7 2020-03-22 Spain 2182 2001. 1804. 2221. 2.94
9 8 2020-03-23 Spain 2696 2533. 2287. 2806. 2.94
10 9 2020-03-24 Spain 3434 3206. 2891. 3555. 2.94
11 10 2020-03-25 Spain 4089 4058. 3648. 4515. 2.94
12 11 2020-03-26 Spain 4858 5136. 4592. 5745. 2.94
13 12 2020-03-27 Spain NA 6501. 5769. 7325. 2.94
14 13 2020-03-28 Spain NA 8228. 7238. 9353. 2.94
15 14 2020-03-29 Spain NA 10414. 9069. 11959. 2.94
16 15 2020-03-30 Spain NA 13181. 11351. 15305. 2.94
17 16 2020-03-31 Spain NA 16682. 14197. 19603. 2.94
18 17 2020-04-01 Spain NA 21115. 17745. 25124. 2.94
19 18 2020-04-02 Spain NA 26724. 22168. 32217. 2.94
dir.create('~/Desktop/ccaa_predictions')
plot_double_time(data, ylog = T) +
labs(subtitle = 'Basic log-linear model weighted at (1 + (1/ days ago)),\nassuming no change in growth trajectory since first day at >150 deaths')
ggsave('~/Desktop/ccaa_predictions/spain.png')
# All ccaas
ccaas <- sort(unique(esp_df$ccaa))
for(i in 1:length(ccaas)){
message(i)
this_ccaa <- ccaas[i]
sub_data <- esp_df %>% mutate(country = ccaa)
try({
data <- make_double_time(
data = sub_data,
the_country = this_ccaa,
n_deaths_start = 5,
time_ahead = 7)
plot_double_time(data, ylog = T) +
labs(subtitle = 'Basic log-linear model weighted at (1 + (1/ days ago)), assuming no change in growth trajectory since first day at >5 deaths')
ggsave(paste0('~/Desktop/ccaa_predictions/',
this_ccaa, '.png'),
height = 4.9,
width = 8.5)
})
}
Error in UseMethod("gather_") :
no applicable method for 'gather_' applied to an object of class "NULL"
Error in UseMethod("gather_") :
no applicable method for 'gather_' applied to an object of class "NULL"
# all_countries <- sort(unique(df_country$country))
# for(i in 1:length(all_countries)){
# this_country <- all_countries[i]
# data <- make_double_time(the_country = this_country, n_deaths_start = 5)
# if(!is.null(data)){
# # print(this_country)
# g <- plot_double_time(data, ylog = F) +
# labs(subtitle = 'Basic log-linear model assuming no change in growth trajectory since first day at >5 deaths')
# ggsave(paste0('~/Desktop/', this_country, '.png'), height = 5, width = 8)
# print(data)
# }
# }
out_list <- curve_list <- list()
counter <- 0
for(i in 1:length(countries)){
message(i)
this_country <- countries[i]
sub_data <-df_country %>% filter(country == this_country)
# Only calculate on countries with n_cases_start or greater cases,
# starting at the first day at n_cases_start or greater
ok <- max(sub_data$deaths, na.rm = TRUE) >= n_deaths_start
if(ok){
counter <- counter + 1
pd <- sub_data %>%
filter(!is.na(deaths)) %>%
mutate(start_date = min(date[deaths >= n_deaths_start])) %>%
mutate(days_since = date - start_date) %>%
filter(days_since >= 0) %>%
mutate(days_since = as.numeric(days_since))
fit <- lm(log(deaths) ~ days_since, data = pd)
# plot(pd$days_since, log(pd$cases))
# abline(fit)
## Slope
# curve <- fit$coef[2]
# Predict days ahead
fake <- tibble(days_since = seq(0, max(pd$days_since) + 5, by = 1))
fake <- left_join(fake, pd %>% dplyr::select(days_since, deaths, date))
fake$predicted <- exp(predict(fit, newdata = fake))
# Doubling time
dt <- log(2)/fit$coef[2]
out <- tibble(country = this_country,
doubling_time = dt)
out_list[[counter]] <- out
curve_list[[counter]] <- fake %>% mutate(country = this_country)
}
}
Error in lm.fit(x, y, offset = offset, singular.ok = singular.ok, ...): NA/NaN/Inf in 'y'
done <- bind_rows(out_list)
curves <- bind_rows(curve_list)
# Get curves back in exponential form
# curves$curve <- exp(curves$curve)
# Join doubling time to curves
joined <- left_join(curves, done)
# Get rid of Italy future (since it's the "leader")
joined <- joined %>%
filter(country != 'Italy' |
date <= (Sys.Date() -1))
# Make long format
long <- joined %>%
dplyr::select(date, days_since, country, deaths, predicted, doubling_time) %>%
tidyr::gather(key, value, deaths:predicted) %>%
mutate(key = Hmisc::capitalize(gsub('_', ' ', key))) %>%
mutate(key = ifelse(key == 'Predicted', 'Predicted (based on current doubling time)', key))
cols <- c('red', 'black')
sub_data <- long %>% filter(country != 'US')
ggplot(data = sub_data,
aes(x = days_since,
y = value,
lty = key,
color = key)) +
geom_line(data = sub_data %>% filter(key != 'Deaths'),
size = 1.2, alpha = 0.8) +
geom_point(data = sub_data %>% filter(key == 'Deaths')) +
geom_line(data = sub_data %>% filter(key == 'Deaths'),
size = 0.8) +
facet_wrap(~paste0(country, '\n',
'(doubling time: ',
round(doubling_time, digits = 1), ' days)'), scales = 'free') +
theme_simple() +
scale_linetype_manual(name ='',
values = c(1,2)) +
scale_color_manual(name = '',
values = cols) +
theme(legend.position = 'top') +
labs(x = 'Days since first day at >5 cumulative deaths',
y = 'Deaths',
title = 'COVID-19 DEATHS: ("predicted" assumes no change in doubling time)',
caption = 'Data from Johns Hopkins. Processing: Joe Brew @joethebrew. Code: github.com/databrew/covid19',
subtitle = '(Doubling time calculated since first day at >5 cumulative deaths)') +
theme(strip.text = element_text(size = 13),
plot.title = element_text(size = 15))
Let’s again overlay Italy.
# Overlay Italy
ol1 <- joined %>% filter(!country %in% 'Italy')
ol2 <- joined %>% filter(country == 'Italy') %>% dplyr::rename(Italy = deaths) %>%
dplyr::select(Italy, days_since)
ol <- left_join(ol1, ol2) %>%
dplyr::select(days_since, date, country, deaths, predicted, Italy,doubling_time)
ol <- tidyr::gather(ol, key, value, deaths: Italy) %>%
mutate(key = Hmisc::capitalize(gsub('_', ' ', key))) %>%
mutate(key = ifelse(key == 'Predicted', 'Predicted (based on current doubling time)', key))
cols <- c('red', 'blue', 'black')
sub_data <- ol %>%
filter(!(key == 'Predicted (based on current doubling time)' &
country == 'Spain' &
days_since > 13))
ggplot(data = sub_data,
aes(x = days_since,
y = value,
lty = key,
color = key)) +
geom_line(data = sub_data %>% filter(!key %in% c('Deaths', 'Italy')),
size = 1.2, alpha = 0.8) +
geom_line(data = sub_data %>% filter(key %in% c('Italy')),
size = 0.8, alpha = 0.8) +
geom_point(data = sub_data %>% filter(key == 'Deaths')) +
geom_line(data = sub_data %>% filter(key == 'Deaths'),
size = 0.8) +
facet_wrap(~paste0(country, '\n',
'(doubling time: ',
round(doubling_time, digits = 1), ' days)'), scales = 'free') +
theme_simple() +
scale_linetype_manual(name ='',
values = c(1,6,2)) +
scale_color_manual(name = '',
values = cols) +
scale_y_log10() +
theme(legend.position = 'top') +
labs(x = 'Days since first day at >5 deaths',
y = 'Deaths',
title = 'COVID-19 DEATHS: ("predicted" assumes no change in doubling time)',
caption = 'Data from Johns Hopkins. Processing: Joe Brew @joethebrew. Code: github.com/databrew/covid19',
subtitle = '(Doubling time calculated since first day at >5 cumulative deaths)') +
theme(strip.text = element_text(size = 13),
plot.title = element_text(size = 15))
Let’s look just at Spain
# Overlay Italy
ol1 <- joined %>% filter(!country %in% 'Italy',
country == 'Spain')
ol2 <- joined %>% filter(country == 'Italy') %>% dplyr::rename(Italy = deaths) %>%
dplyr::select(Italy, days_since)
ol <- left_join(ol1, ol2) %>%
dplyr::select(days_since, date, country, deaths, predicted, Italy,doubling_time)
ol <- tidyr::gather(ol, key, value, deaths: Italy) %>%
mutate(key = Hmisc::capitalize(gsub('_', ' ', key))) %>%
mutate(key = ifelse(key == 'Predicted', 'Predicted (based on current doubling time)',
ifelse(key == 'Deaths', 'Spain', key)))
cols <- c('blue', 'black', 'red')
ggplot(data = ol,
aes(x = days_since,
y = value,
lty = key,
color = key)) +
geom_line(data = ol %>% filter(!key %in% c('Deaths', 'Italy')),
size = 1.2, alpha = 0.8) +
geom_line(data = ol %>% filter(key %in% c('Italy')),
size = 0.8, alpha = 0.8) +
# geom_point(data = ol %>% filter(key == 'Deaths')) +
geom_point(data = ol %>% filter(country == 'Spain',
key == 'Spain'), size = 4, alpha = 0.6) +
geom_line(data = ol %>% filter(key == 'Deaths'),
size = 0.8) +
# facet_wrap(~paste0(country, '\n',
# '(doubling time: ',
# round(doubling_time, digits = 1), ' days)'), scales = 'free') +
theme_simple() +
scale_linetype_manual(name ='',
values = c(1,6,1)) +
scale_color_manual(name = '',
values = cols) +
scale_y_log10() +
theme(legend.position = 'top') +
labs(x = 'Days since first day at >5 deaths',
y = 'Deaths',
title = 'COVID-19 DEATHS: ("predicted" assumes no change in doubling time)',
caption = 'Data from Johns Hopkins. Processing: Joe Brew @joethebrew. Code: github.com/databrew/covid19',
subtitle = '(Doubling time calculated since first day at >5 cumulative deaths)') +
theme(strip.text = element_text(size = 13),
plot.title = element_text(size = 15),
axis.title = element_text(size = 18))
Things are changing very rapidly. And measures being taken by these countries will have an impact on the outbreak.
But it’s important to remember that there is a lag between when an intervention takes place and when its effect is notable. Because of the incubation period - the number of days between someone getting infected and becoming sick - what we do today won’t really have an effect until next weekend. And the clinical cases that present today are among people who got infected a week ago.
Disease control measures work. We can see that clearly in the case of Hubei, Wuhan, Iran, Japan. And they will work in Europe too. But because many of these measures were implemented very recently, we won’t likely see a major effect for at least a few more days.
In the mean time, it’s important to practice social distancing. Stay away from others to keep both you and others safe. Listen to Health Authorities. Take this very seriously.
# Madrid vs Lombardy deaths
n_death_start <- 5
pd <- esp_df %>%
# filter(ccaa == 'Madrid') %>%
dplyr::select(date, ccaa, cases, deaths) %>%
bind_rows(ita %>%
# filter(ccaa == 'Lombardia') %>%
dplyr::select(date, ccaa, cases, deaths)) %>%
arrange(date) %>%
group_by(ccaa) %>%
mutate(first_n_death = min(date[deaths >= n_death_start])) %>%
ungroup %>%
mutate(days_since_n_deaths = date - first_n_death) %>%
filter(is.finite(days_since_n_deaths))
pd$country <- pd$ccaa
pd$cases <- pd$cases
countries <- sort(unique(pd$country))
out_list <- curve_list <- list()
counter <- 0
for(i in 1:length(countries)){
message(i)
this_country <- countries[i]
sub_data <- pd %>% filter(country == this_country)
# Only calculate on countries with n_cases_start or greater cases,
# starting at the first day at n_cases_start or greater
# ok <- max(sub_data$deaths, na.rm = TRUE) >= n_deaths_start
ok <- length(which(sub_data$deaths >= n_deaths_start))
if(ok){
counter <- counter + 1
sub_pd <- sub_data %>%
filter(!is.na(deaths)) %>%
mutate(start_date = min(date[deaths >= n_deaths_start])) %>%
mutate(days_since = date - start_date) %>%
filter(days_since >= 0) %>%
mutate(days_since = as.numeric(days_since))
fit <- lm(log(deaths) ~ days_since, data = sub_pd)
# plot(pd$days_since, log(pd$cases))
# abline(fit)
## Slope
# curve <- fit$coef[2]
# Predict days ahead
fake <- tibble(days_since = seq(0, max(sub_pd$days_since) + 5, by = 1))
fake <- left_join(fake, sub_pd %>% dplyr::select(days_since, deaths, date))
fake$predicted <- exp(predict(fit, newdata = fake))
# Doubling time
dt <- log(2)/fit$coef[2]
out <- tibble(country = this_country,
doubling_time = dt)
out_list[[counter]] <- out
curve_list[[counter]] <- fake %>% mutate(country = this_country)
}
}
done <- bind_rows(out_list)
curves <- bind_rows(curve_list)
# Get curves back in exponential form
# curves$curve <- exp(curves$curve)
# Join doubling time to curves
joined <- left_join(curves, done)
# Make long format
long <- joined %>%
dplyr::select(date, days_since, country, deaths, predicted, doubling_time) %>%
tidyr::gather(key, value, deaths:predicted) %>%
mutate(key = Hmisc::capitalize(gsub('_', ' ', key))) %>%
mutate(key = ifelse(key == 'Predicted', 'Predicted (based on current doubling time)', key))
# Remove those with not enough data to have a doubling time yet
long <- long %>% filter(!is.na(doubling_time))
text_size <- 12
cols <- c('red', 'black')
ggplot(data = long,
aes(x = days_since,
y = value,
lty = key,
color = key)) +
geom_line(data = long %>% filter(key != 'Deaths'),
size = 1.2, alpha = 0.8) +
geom_point(data = long %>% filter(key == 'Deaths')) +
geom_line(data = long %>% filter(key == 'Deaths'),
size = 0.8) +
facet_wrap(~paste0(country, '\n',
'(doubling time: ',
round(doubling_time, digits = 1), ' days)'), scales = 'free') +
theme_simple() +
scale_y_log10() +
scale_linetype_manual(name ='',
values = c(1,2)) +
scale_color_manual(name = '',
values = cols) +
theme(legend.position = 'top') +
labs(x = 'Days since first day at >150 cumulative cases',
y = 'Deaths',
title = 'COVID-19 DEATHS: ("predicted" assumes no change in doubling time)',
caption = 'Data from Johns Hopkins. Processing: Joe Brew @joethebrew. Code: github.com/databrew/covid19',
subtitle = '(Doubling time calculated since first day at >5 cumulative deaths)') +
theme(strip.text = element_text(size = text_size * 0.5),
plot.title = element_text(size = 15))
Let’s overlay Lombardy
# Overlay Italy
ol1 <- joined %>% filter(!country %in% 'Lombardia')
ol2 <- joined %>% filter(country == 'Lombardia') %>% dplyr::rename(Lombardia = deaths) %>%
dplyr::select(Lombardia, days_since)
ol <- left_join(ol1, ol2) %>%
dplyr::select(days_since, date, country, deaths, predicted, Lombardia,doubling_time)
ol <- tidyr::gather(ol, key, value, deaths: Lombardia) %>%
mutate(key = Hmisc::capitalize(gsub('_', ' ', key))) %>%
mutate(key = ifelse(key == 'Predicted', 'Predicted (based on current doubling time)', key))
# Remove those with not enough data to have a doubling time yet
ol <- ol %>% filter(!is.na(doubling_time))
cols <- c('red', 'blue', 'black')
ggplot(data = ol,
aes(x = days_since,
y = value,
lty = key,
color = key)) +
scale_y_log10() +
geom_line(data = ol %>% filter(!key %in% c('Deaths', 'Italy')),
size = 1.2, alpha = 0.8) +
geom_line(data = ol %>% filter(key %in% c('Lombardia')),
size = 0.5, alpha = 0.8) +
geom_point(data = ol %>% filter(key == 'Deaths')) +
geom_line(data = ol %>% filter(key == 'Deaths'),
size = 0.8) +
facet_wrap(~paste0(country, '\n',
'(doubling time: ',
round(doubling_time, digits = 1), ' days)'), scales = 'free') +
theme_simple() +
scale_linetype_manual(name ='',
values = c(1,6,2)) +
scale_color_manual(name = '',
values = cols) +
theme(legend.position = 'top') +
labs(x = 'Days since first day at >5 deaths',
y = 'Deaths',
title = 'COVID-19 DEATHS: ("predicted" assumes no change in doubling time)',
caption = 'Data from Johns Hopkins. Processing: Joe Brew @joethebrew. Code: github.com/databrew/covid19',
subtitle = '(Doubling time calculated since first day at >5 cumulative deaths)') +
theme(strip.text = element_text(size = text_size * 0.5),
plot.title = element_text(size = 15))
Show only Spanish regions vs. Lombardy
text_size <- 14
# Overlay Italy
ol1 <- joined %>% filter(!country %in% 'Lombardia')
ol2 <- joined %>% filter(country == 'Lombardia') %>% dplyr::rename(Lombardia = deaths) %>%
dplyr::select(Lombardia, days_since)
ol <- left_join(ol1, ol2) %>%
dplyr::select(days_since, date, country, deaths, predicted, Lombardia,doubling_time)
ol <- tidyr::gather(ol, key, value, deaths: Lombardia) %>%
mutate(key = Hmisc::capitalize(gsub('_', ' ', key))) %>%
mutate(key = ifelse(key == 'Predicted', 'Predicted (based on current doubling time)', key))
# Remove those with not enough data to have a doubling time yet
ol <- ol %>% filter(!is.na(doubling_time))
# Only Spain
ol <- ol %>% filter(country %in% esp_df$ccaa) %>%
filter(!country %in% 'Aragón')
cols <- c('red', 'blue', 'black')
ggplot(data = ol,
aes(x = days_since,
y = value,
lty = key,
color = key)) +
scale_y_log10() +
geom_line(data = ol %>% filter(!key %in% c('Deaths', 'Lombardia')),
size = 1.2, alpha = 0.8) +
geom_line(data = ol %>% filter(key %in% c('Lombardia')),
size = 0.5, alpha = 0.8) +
geom_point(data = ol %>% filter(key == 'Deaths')) +
geom_line(data = ol %>% filter(key == 'Deaths'),
size = 0.8) +
facet_wrap(~paste0(country, '\n',
'(doubling time: ',
round(doubling_time, digits = 1), ' days)'), scales = 'free') +
theme_simple() +
scale_linetype_manual(name ='',
values = c(1,6,2)) +
scale_color_manual(name = '',
values = cols) +
theme(legend.position = 'top') +
labs(x = 'Days since first day at >5 deaths',
y = 'Deaths',
title = 'COVID-19 DEATHS: ("predicted" assumes no change in doubling time)',
caption = 'Data from Johns Hopkins. Processing: Joe Brew @joethebrew. Code: github.com/databrew/covid19',
subtitle = '(Doubling time calculated since first day at >5 cumulative deaths)') +
theme(strip.text = element_text(size = text_size * 0.6),
plot.title = element_text(size = 15))
Same plot but overlayed
Same as above, but overlaid
text_size <-10
# cols <- c('red', 'black')
long <- long %>% filter(country %in% c('Lombardia',
'Emilia Romagna') |
country %in% esp_df$ccaa) %>%
filter(country != 'Aragón')
places <- sort(unique(long$country))
cols <- colorRampPalette(RColorBrewer::brewer.pal(n = 7, 'Spectral'))(length(places))
cols[which(places == 'Madrid')] <- 'red'
cols[which(places == 'Cataluña')] <- 'purple'
cols[which(places == 'Lombardia')] <- 'darkorange'
cols[which(places == 'Emilia Romagna')] <- 'darkgreen'
long$key <- ifelse(long$key != 'Deaths', 'Predicted', long$key)
long$key <- ifelse(long$key == 'Predicted', 'Muertes\nprevistas',
'Muertes\nobservadas')
# Keep only Madrid, Lombardy, Emilia Romagna
long <- long %>%
filter(country %in% c('Madrid',
'Lombardia',
'Emilia Romagna'))
ggplot(data = long,
aes(x = days_since,
y = value,
lty = key,
color = country)) +
geom_point(data = long %>% filter(key == 'Muertes\nobservadas'), size = 2, alpha = 0.8) +
geom_line(data = long %>% filter(key == 'Muertes\nprevistas'), size = 1, alpha = 0.7) +
geom_line(data = long %>% filter(key != 'Muertes\nprevistas'), size = 0.8) +
theme_simple() +
scale_y_log10() +
scale_linetype_manual(name ='',
values = c(1,4)) +
scale_color_manual(name = '',
values = cols) +
theme(legend.position = 'top') +
# labs(x = 'Days since first day at 5 or more cumulative deaths',
# y = 'Deaths',
# title = 'COVID-19 DEATHS: ("predicted" assumes no change in doubling time)',
# caption = 'Data from Johns Hopkins. Processing: Joe Brew @joethebrew. Code: github.com/databrew/covid19',
# subtitle = '(Doubling time calculated since first day at >5 cumulative deaths)') +
labs(x = 'Días desde el primer día a 5 o más muertes acumuladas',
y = 'Muertes (escala logarítmica)',
title = 'Muertes por COVID-19',
caption = 'Data from Johns Hopkins. Processing: Joe Brew @joethebrew. Code: github.com/databrew/covid19',
subtitle = '(Tasa de crecimiento calculada desde el primer día a 5 o más muertes acumuladas)\n(Muertes "previstas": suponiendo que no hay cambios en la tasa de crecimiento)') +
theme(strip.text = element_text(size = text_size * 0.75),
plot.title = element_text(size = text_size * 3),
legend.text = element_text(size = text_size * 1.5),
axis.title = element_text(size = text_size * 2),
axis.text = element_text(size = text_size * 2))
# cols <- c(cols, 'darkorange')
# ggplot(data = ol,
# aes(x = days_since,
# y = value,
# lty = key,
# color = key)) +
# scale_y_log10() +
# geom_line(aes(color = country)) +
#
# # geom_line(data = ol %>% filter(!key %in% c('Deaths', 'Italy')),
# # size = 1.2, alpha = 0.8) +
# # geom_line(data = ol %>% filter(key %in% c('Lombardia')),
# # size = 0.5, alpha = 0.8) +
# # geom_point(data = ol %>% filter(key == 'Deaths')) +
# # geom_line(data = ol %>% filter(key == 'Deaths'),
# # size = 0.8) +
# theme_simple() +
# scale_linetype_manual(name ='',
# values = c(1,6,2)) +
# scale_color_manual(name = '',
# values = cols) +
# theme(legend.position = 'top') +
# labs(x = 'Days since first day at >5 deaths',
# y = 'Deaths',
# title = 'COVID-19 DEATHS: ("predicted" assumes no change in doubling time)',
# caption = 'Data from Johns Hopkins. Processing: Joe Brew @joethebrew. Code: github.com/databrew/covid19',
# subtitle = '(Doubling time calculated since first day at >5 cumulative deaths)') +
# theme(strip.text = element_text(size = text_size * 1),
# plot.title = element_text(size = 15))
# Map data preparation
if(!'map.RData' %in% dir()){
esp1 <- getData(name = 'GADM', country = 'ESP', level = 1)
# Remove canary
esp1 <- esp1[esp1@data$NAME_1 != 'Islas Canarias',]
espf <- fortify(esp1, region = 'NAME_1')
# Standardize names
# Convert names
map_names <- esp1@data$NAME_1
data_names <- sort(unique(esp_df$ccaa))
names_df <- tibble(NAME_1 = c('Andalucía',
'Aragón',
'Cantabria',
'Castilla-La Mancha',
'Castilla y León',
'Cataluña',
'Ceuta y Melilla',
'Comunidad de Madrid',
'Comunidad Foral de Navarra',
'Comunidad Valenciana',
'Extremadura',
'Galicia',
'Islas Baleares',
'La Rioja',
'País Vasco',
'Principado de Asturias',
'Región de Murcia'),
ccaa = c('Andalucía',
'Aragón',
'Cantabria',
'CLM',
'CyL',
'Cataluña',
'Melilla',
'Madrid',
'Navarra',
'C. Valenciana',
'Extremadura',
'Galicia',
'Baleares',
'La Rioja',
'País Vasco',
'Asturias',
'Murcia'))
espf <- left_join(espf %>% dplyr::rename(NAME_1 = id), names_df)
centroids <- data.frame(coordinates(esp1))
names(centroids) <- c('long', 'lat')
centroids$NAME_1 <- esp1$NAME_1
centroids <- centroids %>% left_join(names_df)
# Get random sampling points
random_list <- list()
for(i in 1:nrow(esp1)){
message(i)
shp <- esp1[i,]
# bb <- bbox(shp)
this_ccaa <- esp1@data$NAME_1[i]
# xs <- runif(n = 500, min = bb[1,1], max = bb[1,2])
# ys <- runif(n = 500, min = bb[2,1], max = bb[2,2])
# random_points <- expand.grid(long = xs, lat = ys) %>%
# mutate(x = long,
# y = lat)
# coordinates(random_points) <- ~x+y
# proj4string(random_points) <- proj4string(shp)
# get ccaa
message('getting locations of randomly generated points')
# polys <- over(random_points,polygons(shp))
# polys <- as.numeric(polys)
random_points <- spsample(shp, n = 20000, type = 'random')
random_points <- data.frame(random_points)
random_points$NAME_1 <- this_ccaa
random_points <- left_join(random_points, names_df) %>% dplyr::select(-NAME_1)
random_list[[i]] <- random_points
}
random_points <- bind_rows(random_list)
random_points <- random_points %>% mutate(long = x,
lat = y)
save(espf,
esp1,
names_df,
centroids,
random_points,
file = 'map.RData')
} else {
load('map.RData')
}
# Define a function for adding zerio
add_zero <-
function (x, n)
{
x <- as.character(x)
adders <- n - nchar(x)
adders <- ifelse(adders < 0, 0, adders)
for (i in 1:length(x)) {
if (!is.na(x[i])) {
x[i] <- paste0(paste0(rep("0", adders[i]), collapse = ""),
x[i], collapse = "")
}
}
return(x)
}
remake_world_map <- FALSE
options(scipen = '999')
if(remake_world_map){
# World map animation
world <- map_data('world')
# world <- ne_countries(scale = "medium", returnclass = "sf")
# Get plotting data
pd <- df_country %>%
dplyr::select(date, lng, lat, n = cases)
dates <- sort(unique(pd$date))
n_days <- length(dates)
# # Define vectors for projection
# vec_lon <- seq(30, -20, length = n_days)
# vec_lat <- seq(25, 15, length = n_days)
dir.create('animation')
for(i in 1:n_days){
message(i, ' of ', n_days)
this_date <- dates[i]
# this_lon <- vec_lon[i]
# this_lat <- vec_lat[i]
# the_crs <-
# paste0("+proj=laea +lat_0=", this_lat,
# " +lon_0=",
# this_lon,
# " +x_0=4321000 +y_0=3210000 +ellps=GRS80 +units=m +no_defs ")
sub_data <- pd %>%
filter(date == this_date)
# coordinates(sub_data) <- ~lng+lat
# proj4string(sub_data) <- proj4string(esp1)
# # sub_data <- spTransform(sub_data,
# # the_crs)
# coordy <- coordinates(sub_data)
# sub_data@data$long <- coordy[,1]
# sub_data@data$lat <- coordy[,2]
g <- ggplot() +
geom_polygon(data = world,
aes(x = long,
y = lat,
group = group),
fill = 'black',
color = 'white',
size = 0.1) +
theme_map() +
geom_point(data = sub_data %>% filter(n > 0) %>% mutate(Deaths = n),
aes(x = lng,
y = lat,
size = Deaths),
color = 'red',
alpha = 0.6) +
geom_point(data = tibble(x = c(0,0), y = c(0,0), Deaths = c(1, 100000)),
aes(x = x,
y = y,
size = Deaths),
color = 'red',
alpha =0.001) +
scale_size_area(name = '', breaks = c(100, 1000, 10000, 100000),
max_size = 25
) +
# scale_size_area(name = '', limits = c(1, 10), breaks = c(0, 10, 30, 50, 70, 100, 200, 500)) +
labs(title = this_date) +
theme(plot.title = element_text(size = 30),
legend.text = element_text(size = 15),
legend.position = 'left')
plot_number <- add_zero(i, 3)
ggsave(filename = paste0('animation/', plot_number, '.png'),
plot = g,
width = 9.5,
height = 5.1)
}
setwd('animation')
system('convert -delay 30x100 -loop 0 *.png result.gif')
setwd('..')
}
make_map <- function(var = 'deaths',
data = NULL,
pop = FALSE,
pop_factor = 100000,
points = FALSE,
line_color = 'white',
add_names = T,
add_values = T,
text_size = 2.7){
if(is.null(data)){
data <- esp_df
}
left <- espf
right <- data[,c('ccaa', paste0(var, '_non_cum'))]
names(right)[ncol(right)] <- 'var'
right <- right %>% group_by(ccaa) %>% summarise(var = sum(var, na.rm = T))
if(pop){
right <- left_join(right, esp_pop)
right$var <- right$var / right$pop * pop_factor
}
map <- left_join(left, right)
if(points){
the_points <- centroids %>%
left_join(right)
g <- ggplot() +
geom_polygon(data = map,
aes(x = long,
y = lat,
group = group),
fill = 'black',
color = line_color,
lwd = 0.4, alpha = 0.8) +
geom_point(data = the_points,
aes(x = long,
y = lat,
size = var),
color = 'red',
alpha = 0.7) +
scale_size_area(name = '', max_size = 20)
} else {
# cols <- c('#008080','#70a494','#b4c8a8','#f6edbd','#edbb8a','#de8a5a','#ca562c')
cols <- RColorBrewer::brewer.pal(n = 8, name = 'Blues')
g <- ggplot(data = map,
aes(x = long,
y = lat,
group = group)) +
geom_polygon(aes(fill = var),
lwd = 0.3,
color = line_color) +
scale_fill_gradientn(name = '',
colours = cols)
# scale_fill_viridis(name = '' ,option = 'magma',
# direction = -1)
}
# Add names?
if(add_names){
centy <- centroids %>% left_join(right)
if(add_values){
centy$label <- paste0(centy$ccaa, '\n(', round(centy$var, digits = 2), ')')
} else {
centy$label <- centy$ccaa
}
g <- g +
geom_text(data = centy,
aes(x = long,
y = lat,
label = label,
group = ccaa),
alpha = 0.7,
size = text_size)
}
g +
theme_map() +
labs(subtitle = paste0('Data as of ', max(data$date))) +
theme(legend.position = 'right')
}
make_dot_map <- function(var = 'deaths',
date = NULL,
pop = FALSE,
pop_factor = 100,
point_factor = 1,
points = FALSE,
point_color = 'darkred',
point_size = 0.6,
point_alpha = 0.5){
if(is.null(date)){
the_date <- max(esp_df$date)
} else {
the_date <- date
}
right <- esp_df[esp_df$date == the_date,c('ccaa', var)]
names(right)[ncol(right)] <- 'var'
if(pop){
right <- left_join(right, esp_pop)
right$var <- right$var / right$pop * pop_factor
}
map_data <- esp1@data %>%
left_join(names_df) %>%
left_join(right)
map_data$var <- map_data$var / point_factor
out_list <- list()
for(i in 1:nrow(map_data)){
sub_data <- map_data[i,]
this_value = round(sub_data$var)
if(this_value >= 1){
this_ccaa = sub_data$ccaa
# get some points
sub_points <- random_points %>% filter(ccaa == this_ccaa)
sampled_points <- sub_points %>% dplyr::sample_n(this_value)
out_list[[i]] <- sampled_points
}
}
the_points <- bind_rows(out_list)
g <- ggplot() +
geom_polygon(data = espf,
aes(x = long,
y = lat,
group = group),
fill = 'white',
color = 'black',
lwd = 0.4, alpha = 0.8) +
geom_point(data = the_points,
aes(x = long,
y = lat),
color = point_color,
size = point_size,
alpha = point_alpha)
g +
theme_map() +
labs(subtitle = paste0('Data as of ', max(esp_df$date)))
}
make_map(var = 'deaths',
points = T) +
labs(title = 'Number of deaths',
caption = '@joethebrew')
make_map(var = 'deaths',
line_color = 'darkgrey',
points = F) +
labs(title = 'Number of deaths',
caption = '@joethebrew')
make_map(var = 'deaths', pop = TRUE, points = T) +
labs(title = 'Number of deaths per 100,000',
caption = '@joethebrew')
make_map(var = 'deaths', pop = TRUE, points = F, line_color = 'darkgrey') +
labs(title = 'Number of deaths per 100,000',
caption = '@joethebrew')
make_dot_map(var = 'deaths', point_size = 0.05) +
labs(title = 'COVID-19 deaths: 1 point = 1 death\nImportant: points are random within each CCAA; do not reflect exact location',
caption = '@joethebrew')
make_map(var = 'cases',
points = T) +
labs(title = 'Number of confirmed cases',
caption = '@joethebrew')
make_map(var = 'cases',
line_color = 'darkgrey',
points = F) +
labs(title = 'Number of confirmed cases',
caption = '@joethebrew')
make_map(var = 'cases', pop = TRUE, points = T) +
labs(title = 'Number of confirmed cases per 100,000',
caption = '@joethebrew')
make_map(var = 'cases', pop = TRUE, points = F,
line_color = 'darkgrey') +
labs(title = 'Number of confirmed cases per 100,000',
caption = '@joethebrew')
make_dot_map(var = 'cases',
point_size = 0.05, point_alpha = 0.5, point_factor = 10) +
labs(title = 'COVID-19 cases: 1 point = 10 cases\nImportant: points are random within each CCAA; do not reflect exact location',
caption = '@joethebrew')